1829627 Members
1689 Online
109992 Solutions
New Discussion

TLB _Datapage error

 
Maurinho
Occasional Contributor

TLB _Datapage error

While loading Dec 2000 patches on a V2250 ; system reboots when completed .
I received " TLB datapage fault , unexpected firmware error has occured . A hard reset is recommended . 'Reset Debug ' may allow to continue . HP Debug :(prompt) ".
I able to continue by doing 'do_reset' from workstation . I can do a clean shutdown ; I only got the error when loading patch bundles .
I think I might need a firmware upgrade ; but
I want to know fi it might be related to Dec 2000 patch bundle .
Any thoughts are welcome ..Thanks
Mauro
2 REPLIES 2
eran maor
Honored Contributor

Re: TLB _Datapage error

Hi

i dont think that this error is becuase of the installing of the bundle but because you install the bundle from the console and saw the error .

from what i m checking all the unexpected firmware error is happening becuase the firmware is not update in the V-class or there is a hardware problem .

i saw 10 case with the this error but not the same error that you got .

to solve the problem i think you shuold open a case in the hp in your country .

love computers
Patrick Wessel
Honored Contributor

Re: TLB _Datapage error

Hello Mauro,
I assume your system didn't write a memory dump, right? (If I'm wrong and the system wrote a dump ask your support partner to read the information in the dump and stop reading my answer)
The TLB is the Transaction Look aside buffer and is responsible for the translation form virtual to real addressing. It's the responsibility of the corresponding driver to load the page directory of the TLB. Due to that are most TLB faults I saw caused by not proper patched drivers. On the other side points the firmware error to a possible hardware problem. You could have checked for an HPMC to verify what the PIM says about the hardware state of the box. But (and this is the start of a sentence with bad news) the behavior of the system shows that the OBP wasn't up and running at the point the problem occurs and the PIM data wasn't copied from the SRAM to the NVRAM. In other words, at the moment you entered the do_reset, the PIM data was gone.
Next time your system suggests you to enter 'reset debug' trust the box and do a 'reset debug'. That will bring you from the debug prompt to the Boot Console Handler (main menu). At this point you can enter PIM to check for a valid time stamp in the PIM data or do a pim_dumper -s from a shell on the SSP (System Support Prosessor - teststation). (The -s option is to read the data out of the SRAM) It will store the information on the SSP in the /spp/data>complex_name> directory.
Nevertheless it's not bad to have the current firmware installed on the system. Verify your firmware version by entering diag_version on the SSP

Good Luck
There is no good troubleshooting with bad data